Bacillary angiomatosis is a newly-discovered disease that occurs primarily among AIDS patients. During 1992, two groups reported that a new organism, Rochalimea henselae caused bacillary angiomatosis and peliosis hepatis, and others reported that this agent was also responsible for persistent bacteremias in immunocompetent patients and for act scratch disease. R. henselae is the same family as the trench fever agent (r. quintana) and is closely related t Bartonella bacilliformis, the agent of Carrion's disease. Indeed, the lesion of bacillary angiomatosis and verruga peruana (B. bacilliformis) are grossly similar and are histologically indistinguishable, both being the result of endothelial hyperplasia and both containing many bacilli in the interstitium. This study is designed to take a first look at the pathogenic capabilities of R. henselae. Because of the great similarities between R. henselae and bartonellae, this study will examine the two organisms in parallel. E. coli will serve as the negative control for all experiments. The study will examine the pathogenesis of bacillary angiomatosis and verruga peruana in three tiers. First, both organisms will be examined for their abilities to attach to and invade cultured endothelial cells. Radioactive techniques as well as light and electron microscopy will be used to locate the sites of endothelial infection. If the bacteria enter the endothelial cells, metabolic inhibitors will be used to characterize the made of entry. Second, the angiogenic potential of both organisms will be examined. Whole living bacteria, killed bacteria, whole cell extracts and culture supernatant fractions will be examined for their abilities to (i) cause in vitro endothelial hyperplasia, and (ii) cause in vivo angiogenesis using a chicken chorioallantoic membrane model. Finally, because endothelial infection may affect endothelial functions other than those directly related to hyperplasia, endothelial activities related to vascular permeability, immunomodulation and thrombosis will be examined for their response to the presence of bartonellae or rochalimeae. The abilities of infected and uninfected endothelial monolayers to produce PGI2, PGE2, TNF-alpha, platelet activating factors, tissue-type plasminogen activator, and thrombomodulin will be examined. These studies should lay the groundwork for further studies to examine molecular and genetic determinants of the abilities of R. henselae and B. bacilliformis to cause angiomatous and systemic disease.